Elsevier

Geochimica et Cosmochimica Acta

Volume 87, 15 June 2012, Pages 243-266
Geochimica et Cosmochimica Acta

The influence of source heterogeneity on the U–Th–Pa–Ra disequilibria in post-glacial tholeiites from Iceland

https://doi.org/10.1016/j.gca.2012.03.041Get rights and content

Abstract

We investigate the relative influence of mantle upwelling velocity and source heterogeneity on the melting rates recorded by 230Th–238U, 231Pa–235U and 226Ra–230Th disequilibria in post-glacial tholeiites from Iceland’s main rift areas. The measured (230Th/238U) ratios range from 1.085 to 1.247, the (231Pa/235U) ratios from 1.333 to 1.925, and the (226Ra/230Th) ratios from 0.801 to 1.218. A general positive correlation between 230Th excesses and distance from the inferred plume centre is consistent with a model of decreasing mantle upwelling velocity with increasing distance from the plume axis. However, the model is not substantiated by the (231Pa/235U) data as the correlation with distance from the plume centre is weak. On the scale of individual eruption centres, the observed U-series are influenced by variations in melt transport time, source porosity, and local variations in mantle upwelling velocity. Broad correlations between (230Th/238U) and (231Pa/235U) and highly incompatible trace element ratios for samples from the Western Volcanic Zone provide, however, evidence for a significant underlying effect of source heterogeneity on the U-series data. Low 230Th and 231Pa excesses in enriched samples from the Western Volcanic Zone with high U/Th, Nb/U and Nb/La indicate that partial melts from an enriched source component, characterised by high melt productivity but low bulk DU/DTh, influence the U-series systematics of the erupted melts. These results re-affirm the presence of comparatively larger abundances of enriched material in the mantle source beneath the South Western Rift of Iceland, which has been suggested based on relationships between highly incompatible element and Pb isotope ratios in Icelandic basalts. Overall, our results highlight the importance of lithological heterogeneity on the melting behaviour of the upper mantle and the composition of oceanic basalts.

Introduction

Uranium series disequilibria in young oceanic basalts provide constraints on the dynamics of partial melting and melt extraction processes in the upper mantle. Specifically, they reveal information about mantle upwelling velocity, melt extraction rate, and residual porosity, but also about intrinsic properties of the mantle source such as its modal mineralogy and melt productivity (e.g., Condomines et al., 1981, Williams and Gill, 1989, McKenzie, 1985, Cohen and O’Nions, 1993, Iwamori, 1993, Spiegelman and Elliott, 1993, Turner et al., 1997, Bourdon et al., 1998, Bourdon et al., 2005, Bourdon et al., 2006, Sims et al., 1999, Stracke et al., 1999, Stracke et al., 2006, Stracke et al., 2003a, Peate et al., 2001, Kokfelt et al., 2003, Lundstrom et al., 2003, Pietruszka et al., 2009, Prytulak and Elliott, 2009).

During partial melting of the upper mantle, differences in residence time of the parent and daughter nuclides in the melt and residual solid mantle cause U-series disequilibrium. Proposed melting models range from ‘dynamic melting’ assuming rapid melt extraction with no equilibration between the partial melts and solid (McKenzie, 1985) to ‘equilibrium porous-flow’ with continuous melt–solid equilibration (Spiegelman and Elliott, 1993). More complex models suggest a so-called two-porosity regime during melt extraction, which implies different degrees of melt–solid equilibration at different depths in the mantle (Iwamori, 1994, Lundstrom et al., 2000, Lundstrom, 2001, Jull et al., 2002). These melting models can be used to explain the observed U-series isotope variation by variation in mantle upwelling velocity, which is directly proportional to the melting rate, and residual porosity during partial melting. Generally, the above-mentioned models make the simplifying assumption that the mantle source has a homogeneous mineralogical composition. The presence of lithological heterogeneity in the mantle, however, may change its melting behaviour, because different lithologies have different modal composition, trace element partitioning characteristics, and melt productivity (Lundstrom et al., 1995, Bourdon et al., 1996, Hirschmann and Stolper, 1996, Stracke et al., 1999, Stracke et al., 2003a, Pertermann and Hirschmann, 2003, Pertermann et al., 2004, Prytulak and Elliott, 2007, Prytulak and Elliott, 2009).

Melt productivity for example – defined as the amount of melt formed per increment of pressure release – is consistently larger for mafic lithologies compared to peridotites (Hirschmann and Stolper, 1996, Asimow et al., 1997, Asimow et al., 2001, Hirschmann et al., 1999a, Kogiso et al., 2004, Pertermann and Hirschmann, 2003) resulting in higher melting rates. Prytulak and Elliott (2009) pointed out that the differences in melt productivity and thus melting rate between peridotitic and pyroxenitic mantle components could have a much larger effect on the U-series nuclides in ocean island basalts (OIB) than differences in mantle upwelling velocity. Russo et al. (2009) suggested that melting of fertile pyroxenite veins could explain the relationships between trace element ratios and 230Th and 226Ra excesses observed in mid ocean ridge basalts (MORB) from the South-East Indian Ridge. The quantitative effect of lithological source heterogeneity on the U-series disequilibria, however, remains difficult to predict due to the uncertainty attached to the relevant melting and partitioning behaviour of lithologically different sources (e.g., Stracke et al., 1999, Bourdon and Sims, 2003).

In Icelandic rocks, correlations between major elements and trace element ratios and long-lived isotopes suggest that melting of at least two components, one isotopically depleted and one isotopically enriched, is required to explain the observed trends (Wood, 1981, Elliott et al., 1991, Maclennan et al., 2003, Stracke et al., 2003b, Kokfelt et al., 2006, Maclennan, 2008a, Stracke and Bourdon, 2009, Peate et al., 2010, Koornneef et al., 2012). Although the nature of the enriched Icelandic source component remains controversial, most previous studies favoured ancient recycled oceanic crust, present in form of small-scale mafic components (e.g., Chauvel and Hémond, 2000, Skovgaard et al., 2001, Stracke et al., 2003b, Kokfelt et al., 2006, Peate et al., 2010). Enrichments in Nb/La and Nb/U ratios combined with higher 206Pb/204Pb in samples from the Western Rift Zone and the Reykjanes Peninsula suggest that the abundance of this enriched component is larger beneath these main rift areas compared to the Northern Rift Zone (Hanan et al., 2000, Koornneef et al., 2012). The inferred lithological source heterogeneity may therefore affect the U-series disequilibria in young Icelandic lavas. Kokfelt et al., 2003, however, mainly attributed the observed differences in (230Th/238U) (where parentheses denote activity ratios) in Icelandic rift-zone lavas to variations in mantle upwelling rate as a result of decreasing mantle potential temperature away from the plume centre (Kokfelt et al., 2003, Bourdon et al., 2006).

Here, we present new 226Ra–230Th–238U and 231Pa–235U disequilibria data on 25 geochemically well characterised post-glacial tholeiites from Iceland’s main rift areas (Koornneef et al., 2012) and replicate analyses of four lavas from Theistareykir previously analysed by Stracke et al., 2006, Stracke et al., 2003a). In addition to 230Th–238U disequilibria, the aim is to use (231Pa/235U) ratios, which are more sensitive to variability in melting rates compared to (230Th/238U) ratios, to evaluate the potential effects of source heterogeneity and the inferred variations in regional upwelling velocity on the U-series disequilibria.

Even though variations in mantle potential temperature have no resolvable influence on the major, trace element, and long-lived isotope systematics (Koornneef et al., 2012), our (230Th/238U) data demonstrate that systematic variation in regional mantle upwelling velocity across Iceland is required (Kokfelt et al., 2006, Bourdon et al., 2006). Variability of the (230Th/238U) and (231Pa/235U) ratios on a local scale and the observed correlations with highly incompatible trace elements reveal an important role of source heterogeneity for establishing the U-series disequilibria in Icelandic rift zone lavas.

Section snippets

Sample preparation and analytical techniques

U, Th, Pa and Ra concentrations and isotope ratios were determined on 25 post glacial tholeiites from Iceland’s main rift areas (Fig. 1). Koornneef et al. (2012) previously reported their major and trace element and Hf and Nd isotope composition. In addition to the samples from the Reykjanes Peninsula (RP, n = 10), the Western Volcanic Zone (WV, n = 7), and the Northern Volcanic Zone (NV, n = 8), we re-analysed four samples from Theistareykir, a small area in the Northern Volcanic Zone, that were

Results

U–Th–Pa and Ra concentrations and 230Th–238U, 231Pa–235U and 226Ra–230Th disequilibria data are presented in Table 1 and Fig. 2, Fig. 3, Fig. 4. Data corrected for post-eruptive decay (Table 1), which is mainly relevant for the 226Ra-disequilibria of the lavas, are also shown in Fig. 3. Note that the accuracy of the age-corrected data is limited by the precision of the available age estimates for each lava flow (Peate et al., 2009, Sinton et al., 2005).

U-series melting models

Several U-series melting models have been proposed (e.g., McKenzie, 1985, Iwamori, 1993, Spiegelman and Elliott, 1993, Lundstrom et al., 1999, Jull et al., 2002). The difference in residence time between the parent and daughter nuclides causes in-growth of the daughter nuclide when the parent nuclide is retained preferentially in the solid residue during partial melting. Since the Iceland lavas show evidence for compositional variation created within melt channels (Maclennan et al., 2007,

The role of crustal processes

Secondary processes such as crystallisation of phases that fractionate the U-series isotopes, radioactive decay during magma storage, and assimilation of hydrothermally altered wall rocks or evolved lavas potentially disturb the melting-induced U-series disequilibria.

U, Th, Pa and Ra are all highly incompatible in olivine, clinopyroxene and plagioclase (Blundy and Wood, 2003, Fabbrizio et al., 2009), which crystallise in the tholeiitic basalts analysed here (Koornneef et al., 2012). Thus their

Conclusions

230Th excesses in recent Icelandic lavas correlate with distance from the plume centre and the mean variation in U-series disequilibria can be explained by mantle upwelling velocities of ∼14 cm/yr at the plume axis to ∼4 cm/yr at the plume periphery. Both the absolute value and the range of inferred upwelling velocities are, however, model-sensitive. The comparatively few (231Pa/235U) data reported here do not substantiate the inferences from the (230Th/238U) data on the effect of the plume. More

Acknowledgements

John Maclennan is thanked for his enthusiasm and help during sample collection in Iceland and for the useful discussions we had during writing up. Julie Prytulak is thanked for providing a thorough review with suggestions that greatly contributed to improvement of the manuscript. Christoph Beier and a third anonymous reviewer are also thanked for their helpful comments. Finally we would like to thank Mark Rehkamper for the editorial handling and helpful additional suggestions and comments. The

References (108)

  • A. Cohen et al.

    Melting rates beneath Hawaii: evidence from uranium series isotopes in recent lavas

    Earth Planet. Sci. Lett.

    (1993)
  • M. Condomines et al.

    230Th–238U disequilibria in historical lavas from Iceland

    Earth Planet. Sci. Lett.

    (1981)
  • L.J. Elkins et al.

    Partitioning of U and Th during garnet pyroxenite partial melting: constraints on the source of alkaline ocean island basalts

    Earth Planet. Sci. Lett.

    (2008)
  • A. Fabbrizio et al.

    Experimental determination of Ra mineral/melt partitioning for feldspars and Ra-226-disequilibrium crystallization ages of plagioclase and alkali-feldspar

    Earth Planet. Sci. Lett.

    (2009)
  • M.A.M. Gee et al.

    Glacioisostacy controls chemical and isotopic characteristics of tholeiites from the Reykjanes peninsula, SW Iceland

    Earth Planet. Sci. Lett.

    (1998)
  • S.J. Goldstein et al.

    231Pa and 230Th chronology of mid-ocean ridge basalts

    Earth Planet. Sci. Lett.

    (1993)
  • E.H. Hauri et al.

    Experimental and natural partitioning of Th, U, Pb and other trace-elements between garnet, clinopyroxene and basaltic melts

    Chem. Geol.

    (1994)
  • G. Ito et al.

    Dynamics of mantle flow and melting at a ridge-centered hotspot: Iceland and the Mid-Atlantic Ridge

    Earth Planet. Sci. Lett.

    (1996)
  • G. Ito et al.

    Mantle flow, melting, and dehydration of the Iceland mantle plume

    Earth Planet. Sci. Lett.

    (1999)
  • H. Iwamori

    238U–230Th–226Ra- and 235U–231Pa disequilibria produced by mantle melting with porous and channel flows

    Earth Planet. Sci. Lett.

    (1994)
  • M. Jull et al.

    Consequences of diffuse and channelled porous melt migration on uranium series disequilibria

    Geochim. Cosmochim. Acta

    (2002)
  • S. Klemme et al.

    Experimental constraints on major and trace element partitioning during partial melting of eclogite

    Geochim. Cosmochim. Acta

    (2002)
  • T.F. Kokfelt et al.

    Upwelling and melting of the Iceland plume from radial variation of 238U–230Th disequilibria in postglacial volcanic rocks

    Earth Planet. Sci. Lett.

    (2003)
  • T.F. Kokfelt et al.

    Time-scales for magmatic differentiation at the Snaefellsjokull central volcano, western Iceland: constraints from U–Th–Pa–Ra disequilibria in post-glacial lavas

    Geochim. Cosmochim. Acta

    (2009)
  • J.M. Koornneef et al.

    A new method for U-Th-Pa-Ra separation and accurate measurement of 234U–230Th231Pa–226Ra disequilibria in volcanic rocks by MC-ICPMS

    Chem. Geol.

    (2010)
  • T. Kuritani et al.

    Geochemical evolution of historical lavas from Askja Volcano, Iceland: implications for mechanisms and timescales of magmatic differentiation

    Geochim. Cosmochim. Acta

    (2011)
  • T.Z. LaTourrette et al.

    Experimental-determination of U-partitioning and Th-partitioning between clinopyroxene and natural and synthetic basaltic liquid

    Earth Planet. Sci. Lett.

    (1992)
  • C.C. Lundstrom et al.

    Investigating solid mantle upwelling rates beneath mid-ocean ridges using U-series disequilibria, 1: a global approach

    Earth Planet. Sci. Lett.

    (1998)
  • C.C. Lundstrom et al.

    A geochemically consistent hypothesis for MORB generation

    Chem. Geol.

    (2000)
  • C.C. Lundstrom et al.

    U-series disequilibria in volcanic rocks from the Canary Islands: plume versus lithospheric melting

    Geochim. Cosmochim. Acta

    (2003)
  • J. Maclennan

    Lead isotope variability in olivine-hosted melt inclusions from Iceland

    Geochim. Cosmochim. Acta

    (2008)
  • J. Maclennan et al.

    Plume-driven upwelling under central Iceland

    Earth Planet. Sci. Lett.

    (2001)
  • P. McDade et al.

    Trace element partitioning on the Tinaquillo Lherzolite solidus at 1.5 GPa

    Phys. Earth Planet. Inter.

    (2003)
  • D. McKenzie

    230Th–238U disequilibrium and the melting processes beneath ridge axes

    Earth Planet. Sci. Lett.

    (1985)
  • D. McKenzie et al.

    Source enrichment processes responsible for isotopic anomalies in oceanic island basalts

    Geochim. Cosmochim. Acta

    (2004)
  • D.W. Peate et al.

    238U–230Th constraints on mantle upwelling and plume-ridge interaction along the Reykjanes Ridge

    Earth Planet. Sci. Lett.

    (2001)
  • D.A. Pickett et al.

    Observations of 231Pa/235U disequilibrium in volcanic rocks

    Earth Planet. Sci. Lett.

    (1997)
  • J. Prytulak et al.

    TiO2 enrichment in ocean island basalts

    Earth Planet. Sci. Lett.

    (2007)
  • J. Prytulak et al.

    Determining melt productivity of mantle sources from 238U–230Th- and 235U–231Pa disequilibria; an example from Pico Island, Azores

    Geochim. Cosmochim. Acta

    (2009)
  • O. Reubi et al.

    Assimilation of the plutonic roots of the Andean arc controls variations in U-series disequilibria at Volcan Llaima, Chile

    Earth Planet. Sci. Lett.

    (2011)
  • N.M. Ribe et al.

    The dynamics of plume-ridge interaction.1. Ridge-centered plumes

    Earth Planet. Sci. Lett.

    (1995)
  • C.J. Russo et al.

    Mantle melting and magma supply to the Southeast Indian Ridge: the roles of lithology and melting conditions from U-series disequilibria

    Earth Planet. Sci. Lett.

    (2009)
  • V.J.M. Salters et al.

    Trace element partitioning during the initial stages of melting beneath mid-ocean ridges

    Earth Planet. Sci. Lett.

    (1999)
  • O. Sigmarsson et al.

    Mantle and crustal contribution in the genesis of recent basalts from off-rift zones in Iceland: constraints from Th, Sr and O isotopes

    Earth Planet. Sci. Lett.

    (1992)
  • K.W.W. Sims et al.

    Porosity of the melting zone and variations in the solid mantle upwelling rate beneath Hawaii: inferences from 238U–230Th–226Ra and 235U–231Pa disequilibria

    Geochim. Cosmochim. Acta

    (1999)
  • K.W.W. Sims et al.

    Chemical and isotopic constraints on the generation and transport of magma beneath the East Pacific Rise

    Geochim. Cosmochim. Acta

    (2002)
  • A.C. Skovgaard et al.

    Osmium–oxygen isotopic evidence for a recycled and strongly depleted component in the Iceland mantle plume

    Earth Planet. Sci. Lett.

    (2001)
  • M. Spiegelman et al.

    Consequences of melt transport for uranium series disequilibrium in young lavas

    Earth Planet. Sci. Lett.

    (1993)
  • A. Stracke et al.

    The importance of melt extraction for tracing mantle heterogeneity

    Geochim. Cosmochim. Acta

    (2009)
  • A. Stracke et al.

    Melt extraction in the Earth’s mantle: constraints from U–Th–Pa–Ra studies in oceanic basalts

    Earth Planet. Sci. Lett.

    (2006)
  • Cited by (24)

    • Spatial distribution and geochemical characterization of Icelandic mantle end-members: Implications for plume geometry and melting processes

      2022, Chemical Geology
      Citation Excerpt :

      Our forward modelling demonstrates that even the samples that did not yield PRIMELT solutions have compositions that can be generated from a melting region that is consistent with the PRIMELT results. Calculated melt fractions and their relations to geochemical parameters demonstrate that melting processes, which are consistent with modulation by lithosphere thickness in Iceland (Kokfelt et al., 2006; Peate et al., 2010; Koornneef et al., 2012b; Niu, 2021), have a significant impact on how distinct mantle components in the Icelandic mantle are extracted (Fig. 10). Furthermore, calculated mantle TP show that TP variations have a largely insignificant role in generating geochemical heterogeneity in Icelandic volcanic rocks.

    • Testing pyroxenite versus peridotite sources for marine basalts using U-series isotopes

      2019, Lithos
      Citation Excerpt :

      Due to their mostly subaerial exposures and historical records, many ocean island basalts are age-constrained by historic data and stratigraphy, making initial 230Th-226Ra and 235U-231Pa disequilibria at eruption well-characterized (see Fig. 4, and references therein). However, passing through thicker crust and lithosphere can result in greater degrees of crustal assimilation, as observed at certain Azores islands, Reunion, Iceland, and the Canaries (Bourdon et al., 1998; Claude-Ivanaj et al., 1998; Koornneef et al., 2012; Lundstrom, 2003; Pietruszka et al., 2009; Prytulak et al., 2014; Turner et al., 2017). Taking the above considerations into account, U-series disequilibrium studies of OIB have provided extensive evidence for mantle upwelling rate variations at ocean island settings, with or without involvement of a second lithology.

    • Rift–plume interaction reveals multiple generations of recycled oceanic crust in Azores lavas

      2017, Geochimica et Cosmochimica Acta
      Citation Excerpt :

      Mafic mineralogies (pyroxenite, eclogite) have a lower solidus and a higher melt productivity than peridotite (Pertermann and Hirschmann, 2003; Ito and Mahoney, 2005; Prytulak and Elliott, 2009; Bizimis and Peslier, 2015). Koornneef et al. (2012) emphasized the significant effect of lithological heterogeneity on melt productivity in oceanic basalts on the scale of individual eruption centers, based on a relationship between (230Th/238U) and (231Pa/235U) and incompatible trace element ratios in Icelandic basalts. Prytulak and Elliott (2009) calculated melt productivity in Pico using the same method and concluded measured values are consistent with a peridotitic source, albeit independent evidence for a mafic component in the source of Pico lavas (Moreira et al., 1999; see also Millet et al., 2009; Madureira et al., 2011; this study).

    View all citing articles on Scopus
    1

    Present address: Institut für Mineralogie, Westfälische Wilhelms Universität, 48149 Münster, Germany.

    2

    Present address: Laboratoire de Géologie de Lyon, Ecole Normale Supérieure de Lyon, UCBL and CNRS, Lyon, France.

    View full text